Improved documentation for DataFrame.join #12193
Conversation
| * left: use calling frame's index or column(s) | ||
| * right: use other frame's index | ||
| * outer: form union of calling frame's index or column(s) with | ||
| other frame's index |
There was a problem hiding this comment.
I think this is more confusing, as this depends if on is specified or not (so you can simply say that)
There was a problem hiding this comment.
What about this?
* left: use calling frame's index (or column if on is specified)
* right: use other frame's index
* outer: form union of calling frame's index (or column if on is specified) with other frame's index
|
can you update |
DOC: improves DataFrame.join documentation
|
Done, sorry for the delay. |
|
|
||
| Perform a left join using caller's key column and other frame's index | ||
|
|
||
| >>> caller.join(other.set_index('key'), on='key', how='left', |
There was a problem hiding this comment.
can you add this same example w/o using .set_index as well. (and w/o on), and indicate the difference between them.
There was a problem hiding this comment.
Just add this example? caller.join(other, how='left', lsuffix='_l', rsuffix='_r')
There was a problem hiding this comment.
@jreback I don't think it is possible to not use set_index, as join always uses the index of other (which is actually really confusing ...)
@edublancas the lsuffix='_l', rsuffix='_r' is redundant in this case, so I would leave it out
There was a problem hiding this comment.
What about having just these two examples:
Perform a left join using caller's key column and other frame's index
caller.join(other.set_index('key'), on='key', how='left')
Set key as the index column on caller and other, then perform an index-on-index join.
caller.set_index('key').join(other.set_index('key'), how='left')
|
pls rebase/update |
|
Sorry for the delay, I've been working to meet some deadlines for a project. I'll update in the next few days. |
|
can you rebase/update |
| index-on-index and index-on-column(s) joins, but *joins on indexes* by default | ||
| rather than trying to join on common columns (the default behavior for | ||
| ``merge``). If you are joining on index, you may wish to use ``DataFrame.join`` | ||
| to save yourself some typing. |
There was a problem hiding this comment.
Why did you shorten this?
I think the " joins on indexes by default" is very useful explanation
There was a problem hiding this comment.
I think the shorter explanation is better:
index-on-index (by default) and column(s)-on-index join. If you are joining on index only, you may wish to use
DataFrame.jointo save yourself some typing.
|
can you update according to comments |
|
can you rebase / update? |
|
I think the documentation is clear now. There are 3 examples, one using the dataframes original indexes and two joining using the key columns, the first one setting key as the index in both and the second one using on. |
|
thanks @edublancas nice improvement! |
closes #12188
I modified the description in DataFrame.join to make clear the difference with DataFrame.merge, also added examples.